{ "results": { "original_capability_instruct": { "exact_match,strict-match": 0.5630801459168563, "exact_match_stderr,strict-match": 0.0028483348465514185, "alias": "original_capability_instruct" }, "meta_arc_0shot_instruct": { "alias": " - meta_arc_0shot_instruct", "exact_match,strict-match": 0.8248927038626609, "exact_match_stderr,strict-match": 0.01113972223585952 }, "meta_gpqa_0shot_cot_instruct": { "alias": " - meta_gpqa_0shot_cot_instruct", "exact_match,strict-match": 0.296875, "exact_match_stderr,strict-match": 0.021609729061250887 }, "meta_mmlu_0shot_instruct": { "alias": " - meta_mmlu_0shot_instruct", "exact_match,strict-match": 0.6815980629539952, "exact_match_stderr,strict-match": 0.003931452244804845 }, "meta_mmlu_pro_5shot_instruct": { "alias": " - meta_mmlu_pro_5shot_instruct", "exact_match,strict-match": 0.4093251329787234, "exact_match_stderr,strict-match": 0.004482884901882547 } }, "groups": { "original_capability_instruct": { "exact_match,strict-match": 0.5630801459168563, "exact_match_stderr,strict-match": 0.0028483348465514185, "alias": "original_capability_instruct" } }, "group_subtasks": { "original_capability_instruct": [ "meta_arc_0shot_instruct", "meta_gpqa_0shot_cot_instruct", "meta_mmlu_0shot_instruct", "meta_mmlu_pro_5shot_instruct" ] }, "configs": { "meta_arc_0shot_instruct": { "task": "meta_arc_0shot_instruct", "dataset_path": "meta-llama/llama-3.1-8_b-instruct-evals", "dataset_name": "Llama-3.1-8B-Instruct-evals__arc_challenge__details", "test_split": "latest", "process_docs": "def process_docs(dataset: datasets.Dataset) -> datasets.Dataset:\n def _process_doc(doc: dict) -> dict:\n out_doc = {\n \"problem\": doc[\"input_question\"],\n \"gold\": doc[\"input_correct_responses\"][0],\n }\n return out_doc\n dataset = dataset.select_columns([\"input_question\", \"input_correct_responses\", \"input_final_prompts\", \"is_correct\",\"input_question_hash\",\"input_choice_list\",\"output_prediction_text\"])\n dataset = dataset.rename_column(\"is_correct\",\"previously_is_correct\")\n dataset = dataset.map(_process_doc)\n return dataset.map(_process_doc)\n", "doc_to_text": "def doc_to_text(doc: dict) -> str:\n return doc[\"input_final_prompts\"][0]\n", "doc_to_target": "gold", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "exact_match", "aggregation": "mean", "higher_is_better": true, "ignore_case": true, "ignore_punctuation": true } ], "output_type": "generate_until", "generation_kwargs": { "until": [], "do_sample": false, "temperature": 0.0, "max_gen_toks": 2048 }, "repeats": 1, "filter_list": [ { "name": "strict-match", "filter": [ { "function": "regex", "group_select": -1, "regex_pattern": "([A-Z])" }, { "function": "take_first" } ] } ], "should_decontaminate": false, "metadata": { "version": 1.0 } }, "meta_gpqa_0shot_cot_instruct": { "task": "meta_gpqa_0shot_cot_instruct", "dataset_path": "meta-llama/llama-3.1-8_b-instruct-evals", "dataset_name": "Llama-3.1-8B-Instruct-evals__gpqa__details", "test_split": "latest", "process_docs": "def process_docs(dataset: datasets.Dataset) -> datasets.Dataset:\n def _process_doc(doc: dict) -> dict:\n out_doc = {\n \"problem\": doc[\"input_question\"],\n \"gold\": doc[\"input_correct_responses\"][0],\n }\n return out_doc\n dataset = dataset.select_columns([\"input_question\", \"input_correct_responses\", \"input_final_prompts\", \"is_correct\",\"input_question_hash\",\"input_choice_list\",\"output_prediction_text\"])\n dataset = dataset.rename_column(\"is_correct\",\"previously_is_correct\")\n dataset = dataset.map(_process_doc)\n return dataset.map(_process_doc)\n", "doc_to_text": "def doc_to_text(doc: dict) -> str:\n return doc[\"input_final_prompts\"][0]\n", "doc_to_target": "gold", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "exact_match", "aggregation": "mean", "higher_is_better": true, "ignore_case": true, "ignore_punctuation": true } ], "output_type": "generate_until", "generation_kwargs": { "until": [], "do_sample": false, "temperature": 0.0, "max_gen_toks": 2048 }, "repeats": 1, "filter_list": [ { "name": "strict-match", "filter": [ { "function": "regex", "group_select": -1, "regex_pattern": "best answer is ([A-Z])" }, { "function": "take_first" } ] } ], "should_decontaminate": false, "metadata": { "version": 1.0 } }, "meta_mmlu_0shot_instruct": { "task": "meta_mmlu_0shot_instruct", "dataset_path": "meta-llama/llama-3.1-8_b-instruct-evals", "dataset_name": "Llama-3.1-8B-Instruct-evals__mmlu__0_shot__cot__details", "test_split": "latest", "process_docs": "def process_docs_instruct(dataset: datasets.Dataset) -> datasets.Dataset:\n def _process_doc(doc: dict) -> dict:\n out_doc = {\n \"problem\": doc[\"input_question\"],\n \"gold\": doc[\"input_correct_responses\"][0],\n }\n return out_doc\n dataset = dataset.select_columns([\"input_question\", \"input_correct_responses\", \"input_final_prompts\", \"is_correct\",\"input_question_hash\",\"input_choice_list\",\"output_prediction_text\"])\n dataset = dataset.rename_column(\"is_correct\",\"previously_is_correct\")\n dataset = dataset.map(_process_doc)\n return dataset.map(_process_doc)\n", "doc_to_text": "def doc_to_text_instruct(doc: dict) -> str:\n return doc[\"input_final_prompts\"][0]\n", "doc_to_target": "gold", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "exact_match", "aggregation": "mean", "higher_is_better": true, "ignore_case": true, "ignore_punctuation": true } ], "output_type": "generate_until", "generation_kwargs": { "until": [], "do_sample": false, "temperature": 0.0, "max_gen_toks": 1024 }, "repeats": 1, "filter_list": [ { "name": "strict-match", "filter": [ { "function": "regex", "group_select": -1, "regex_pattern": "best answer is ([A-Z])" }, { "function": "take_first" } ] } ], "should_decontaminate": false, "metadata": { "version": 1.0 } }, "meta_mmlu_pro_5shot_instruct": { "task": "meta_mmlu_pro_5shot_instruct", "dataset_path": "meta-llama/llama-3.1-8_b-instruct-evals", "dataset_name": "Llama-3.1-8B-Instruct-evals__mmlu_pro__details", "test_split": "latest", "process_docs": "def meta_process_docs(dataset: datasets.Dataset) -> datasets.Dataset:\n def _process_doc(doc: dict) -> dict:\n out_doc = {\n \"problem\": doc[\"input_question\"],\n \"gold\": doc[\"input_correct_responses\"][0],\n }\n return out_doc\n dataset = dataset.select_columns([\"input_question\", \"input_correct_responses\", \"input_final_prompts\", \"is_correct\",\"input_question_hash\",\"input_choice_list\",\"output_prediction_text\"])\n dataset = dataset.rename_column(\"is_correct\",\"previously_is_correct\")\n dataset = dataset.map(_process_doc)\n return dataset.map(_process_doc)\n", "doc_to_text": "def meta_doc_to_text(doc: dict) -> str:\n return doc[\"input_final_prompts\"][0]\n", "doc_to_target": "gold", "description": "", "target_delimiter": " ", "fewshot_delimiter": "\n\n", "num_fewshot": 0, "metric_list": [ { "metric": "exact_match", "aggregation": "mean", "higher_is_better": true, "ignore_case": true, "ignore_punctuation": true } ], "output_type": "generate_until", "generation_kwargs": { "until": [], "do_sample": false, "temperature": 0.0, "max_gen_toks": 1024 }, "repeats": 1, "filter_list": [ { "name": "strict-match", "filter": [ { "function": "regex", "group_select": -1, "regex_pattern": "best answer is ([A-Z])" }, { "function": "take_first" } ] } ], "should_decontaminate": false, "metadata": { "version": 1.0 } } }, "versions": { "meta_arc_0shot_instruct": 1.0, "meta_gpqa_0shot_cot_instruct": 1.0, "meta_mmlu_0shot_instruct": 1.0, "meta_mmlu_pro_5shot_instruct": 1.0 }, "n-shot": { "meta_arc_0shot_instruct": 0, "meta_gpqa_0shot_cot_instruct": 0, "meta_mmlu_0shot_instruct": 0, "meta_mmlu_pro_5shot_instruct": 0 }, "higher_is_better": { "meta_arc_0shot_instruct": { "exact_match": true }, "meta_gpqa_0shot_cot_instruct": { "exact_match": true }, "meta_mmlu_0shot_instruct": { "exact_match": true }, "meta_mmlu_pro_5shot_instruct": { "exact_match": true }, "original_capability_instruct": { "exact_match": true } }, "n-samples": { "meta_arc_0shot_instruct": { "original": 1165, "effective": 1165 }, "meta_gpqa_0shot_cot_instruct": { "original": 448, "effective": 448 }, "meta_mmlu_0shot_instruct": { "original": 14042, "effective": 14042 }, "meta_mmlu_pro_5shot_instruct": { "original": 12032, "effective": 12032 } }, "config": { "model": "LlamaPlusWrapper", "model_args": { "pretrained": "/shared/user/fine-tune/coach/model/llama3-10b-hf-checkpoint-sft-padding-opencoder-openmath-train-concat-no-divloss-dlerp-16-from-scratch-v4/hf/checkpoint-11000", "dtype": "bfloat16", "tensor_parallel_size": 1, "gpu_memory_utilization": 0.5, "data_parallel_size": 8, "max_model_len": 8192, "cpu_offload_gb": 0, "enable_prefix_caching": false, "add_bos_token": true, "seed": 42, "register_model": "/shared/user/fine-tune/coach/model/llama3-10b-hf-checkpoint-sft-padding-opencoder-openmath-train-concat-no-divloss-dlerp-16-from-scratch-v4/hf/checkpoint-11000" }, "batch_size": 8, "batch_sizes": [], "device": null, "use_cache": null, "limit": null, "bootstrap_iters": 100000, "gen_kwargs": null, "random_seed": 0, "numpy_seed": 1234, "torch_seed": 1234, "fewshot_seed": 1234 }, "git_hash": null, "date": 1735113856.3120685, "pretty_env_info": "PyTorch version: 2.4.0+cu118\nIs debug build: False\nCUDA used to build PyTorch: 11.8\nROCM used to build PyTorch: N/A\n\nOS: CBL-Mariner/Linux (x86_64)\nGCC version: (GCC) 11.2.0\nClang version: Could not collect\nCMake version: version 3.21.4\nLibc version: glibc-2.35\n\nPython version: 3.10.14 (main, Jul 14 2024, 22:24:12) [GCC 11.2.0] (64-bit runtime)\nPython platform: Linux-5.15.164.1-1.cm2-x86_64-with-glibc2.35\nIs CUDA available: True\nCUDA runtime version: 11.8.89\nCUDA_MODULE_LOADING set to: LAZY\nGPU models and configuration: \nGPU 0: NVIDIA A100-SXM4-80GB\nGPU 1: NVIDIA A100-SXM4-80GB\nGPU 2: NVIDIA A100-SXM4-80GB\nGPU 3: NVIDIA A100-SXM4-80GB\nGPU 4: NVIDIA A100-SXM4-80GB\nGPU 5: NVIDIA A100-SXM4-80GB\nGPU 6: NVIDIA A100-SXM4-80GB\nGPU 7: NVIDIA A100-SXM4-80GB\n\nNvidia driver version: 550.54.15\ncuDNN version: Probably one of the following:\n/usr/lib/libcudnn.so.8.9.5\n/usr/lib/libcudnn_adv_infer.so.8.9.5\n/usr/lib/libcudnn_adv_train.so.8.9.5\n/usr/lib/libcudnn_cnn_infer.so.8.9.5\n/usr/lib/libcudnn_cnn_train.so.8.9.5\n/usr/lib/libcudnn_ops_infer.so.8.9.5\n/usr/lib/libcudnn_ops_train.so.8.9.5\nHIP runtime version: N/A\nMIOpen runtime version: N/A\nIs XNNPACK available: True\n\nCPU:\nArchitecture: x86_64\nCPU op-mode(s): 32-bit, 64-bit\nAddress sizes: 48 bits physical, 48 bits virtual\nByte Order: Little Endian\nCPU(s): 256\nOn-line CPU(s) list: 0-255\nVendor ID: AuthenticAMD\nModel name: AMD EPYC 7763 64-Core Processor\nCPU family: 25\nModel: 1\nThread(s) per core: 2\nCore(s) per socket: 64\nSocket(s): 2\nStepping: 1\nFrequency boost: enabled\nCPU max MHz: 3529.0520\nCPU min MHz: 1500.0000\nBogoMIPS: 4899.59\nFlags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 invpcid_single hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 invpcid cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold v_vmsave_vmload vgif v_spec_ctrl umip pku ospke vaes vpclmulqdq rdpid overflow_recov succor smca\nVirtualization: AMD-V\nL1d cache: 4 MiB (128 instances)\nL1i cache: 4 MiB (128 instances)\nL2 cache: 64 MiB (128 instances)\nL3 cache: 512 MiB (16 instances)\nNUMA node(s): 8\nNUMA node0 CPU(s): 0-15,128-143\nNUMA node1 CPU(s): 16-31,144-159\nNUMA node2 CPU(s): 32-47,160-175\nNUMA node3 CPU(s): 48-63,176-191\nNUMA node4 CPU(s): 64-79,192-207\nNUMA node5 CPU(s): 80-95,208-223\nNUMA node6 CPU(s): 96-111,224-239\nNUMA node7 CPU(s): 112-127,240-255\nVulnerability Gather data sampling: Not affected\nVulnerability Itlb multihit: Not affected\nVulnerability L1tf: Not affected\nVulnerability Mds: Not affected\nVulnerability Meltdown: Not affected\nVulnerability Mmio stale data: Not affected\nVulnerability Reg file data sampling: Not affected\nVulnerability Retbleed: Not affected\nVulnerability Spec rstack overflow: Mitigation; safe RET, no microcode\nVulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp\nVulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization\nVulnerability Spectre v2: Mitigation; Retpolines; IBPB conditional; IBRS_FW; STIBP always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected\nVulnerability Srbds: Not affected\nVulnerability Tsx async abort: Not affected\n\nVersions of relevant libraries:\n[pip3] flake8==7.1.1\n[pip3] flash-attn==2.6.3+cu118torch2.4cxx11abifalse\n[pip3] mypy-extensions==1.0.0\n[pip3] numpy==1.24.3\n[pip3] torch==2.4.0+cu118\n[pip3] torch-tb-profiler==0.4.1\n[pip3] torchsummary==1.5.1\n[pip3] torchvision==0.19.0+cu118\n[pip3] triton==3.0.0\n[conda] Could not collect", "transformers_version": "4.46.2", "upper_git_hash": null, "tokenizer_pad_token": [ "", "128256" ], "tokenizer_eos_token": [ "<|eot_id|>", "128009" ], "tokenizer_bos_token": [ "<|begin_of_text|>", "128000" ], "eot_token_id": 128009, "max_length": 8192 }